OUP user menu

Cloning and characterization of the Burkholderia vietnamiensis norM gene encoding a multi-drug efflux protein

Christine C. Fehlner-Gardiner, Miguel A. Valvano
DOI: http://dx.doi.org/10.1111/j.1574-6968.2002.tb11403.x 279-283 First published online: 1 October 2002


Polymyxin B-sensitive mutants in Burkholderia vietnamiensis (Burkholderia cepacia genomovar V) were generated with a mini-Tn5 encoding tetracycline resistance. One of the transposon mutants had an insertion in the norM gene encoding a multi-drug efflux protein. Expression of B. vietnamiensis norM in an Escherichia coli acrAB deletion mutant complemented its norfloxacin hypersensitivity, indicating that the protein functions in drug efflux. However, no effect on antibiotic sensitivity other than sensitivity to polymyxin B was observed in the B. vietnamiensis norM mutant. We demonstrate that increased polymyxin sensitivity in B. vietnamiensis was associated with the presence of tetracycline in the growth medium, a phenotype that was partially suppressed by expression of the norM gene.

  • Cationic peptide
  • Antibiotic resistance
  • Polymyxin
  • Tetracycline
  • Efflux pump
  • Burkholderia cepacia

1 Introduction

Burkholderia cepacia is a Gram-negative bacterium that has become an important opportunistic pathogen for individuals with cystic fibrosis (CF) and chronic granulomatous disease[1]. Phenotypically similar isolates originally classified as B. cepacia have been reclassified into distinct genomovars that form what is collectively known as the B. cepacia complex[2]. In some cystic fibrosis patients, lung infection with B. cepacia complex strains may result in a rapidly fatal, necrotizing pneumonia known as cepacia syndrome[1]. B. cepacia isolates are usually resistant to most clinically useful antibiotics including cationic peptides. The inherent antibiotic resistance of B. cepacia may be attributed to reduced outer membrane permeability 3,4], production of modifying enzymes such as β-lactamases[5], alteration of antibiotic targets[6], and the ability to survive intracellularly [710].

Active efflux processes represent major antibiotic resistance mechanisms in many bacteria[11]. Five classes of efflux systems have been described. They include the resistance-nodulation-division (RND) family, the major facilitator superfamily (MFS), the small multi-drug resistance (SMR) family, the ATP-binding cassette (ABC) family, and the recently identified multi-drug and toxic compound extrusion (MATE) family 12,13]. In Gram-negative bacteria, antibiotic efflux and low outer membrane permeability work in concert to provide high levels of resistance, particularly in the case of RND and MFS systems which pump antibiotics across both membranes[11]. Multi-drug efflux pumps of the RND and MFS types have been described in cystic fibrosis isolates of B. cepacia14,15]. Here we describe the discovery of a gene encoding a MATE-type efflux protein from Burkholderia vietnamiensis (formerly B. cepacia genomovar V) and present evidence supporting a role for this protein in resistance to the cationic peptide polymyxin B (PMB).

2 Materials and methods

2.1 Bacterial strains and plasmids

B. vietnamiensis CEP040 is a genomovar V isolate obtained from a CF patient[7]. Other B. cepacia complex isolates used were FC124 (genomovar I), CEP484 (Burkholderia multivorans, formerly genomovar II), CEP511 (genomovar III), FC473 (Burkholderia stabilis, formerly genomovar IV), and CEP021 (genomovar VI). Escherichia coli DH5α was used as a host for cloning experiments. Bacteria were grown in Luria–Bertani (LB) medium. Medium for growth of B. cepacia was supplemented, as required, with 100 μg ml−1 tetracycline (Tc) and/or 100 μg ml−1 trimethoprim (Tp) and that for growth of E. coli was supplemented with 20 μg ml−1 Tc, 50 μg ml−1−1 Tp, 40 μg ml−1 kanamycin (Km), or 100 μg ml−1 ampicillin (Ap). The plasmids used in this study are listed in Table 1.

View this table:

Plasmids used in this study

PlasmidCharacteristicsSource or Ref.
pTnMod-OTcSelf-cloning mini-Tn5 derivative, TcR[16]
pEX1Expression vector, ApR[23]
pEX1-norMFull-length norM cloned into pEX1This study
pKK223–3Expression vector, ApR[24]
pCFG1pKK223-3, Ptac was replaced with Pdhfr from pTnMod-OTp′, ApRThis study
pTnMod-OTp′Self-cloning mini-Tn5 derivative, TpR[16]
pME6000Cloning vector, TcRS. Heeb
pScosBC1E. coliBurkholderia shuttle cosmid vector, ApR, TpRE. Mahenthiralingam
pME6000TppME6000 carrying the dhfr gene from pScosBC1, TcR, TpRThis study
pMETpΔTetpME6000Tp digested with Eco RV and religated to remove tetA, TpRThis study
pMETpΔTet-norMpMETpΔTet with full-length norM, preceded by Pdhfr, cloned into the Bam HI siteThis study
pRK2013RK2 derivative used as helper for conjugations, KmR, Mob+, Tra+, ColE1[25]

2.2 Isolation of B. vietnamiensis transposon mutants

B. vietnamiensis CEP040 PMB-sensitive mutants were generated by transposon mutagenesis with the mini-Tn5 derivative pTnMod-OTc, which was introduced into CEP040 by conjugation[16]. Transposon mutants were selected by plating the exconjugants on Vogel–Bonner minimal agar containing Tc and 5 μg ml−1 gentamicin. The resulting Tc-resistant colonies were replica plated onto LB–Tc plates containing 60 μg ml−1 PMB sulfate (Sigma Chemical Co., St. Louis, MO, USA). Genomic DNA flanking the transposon insertion in mutants that failed to grow in the presence of PMB was directly cloned and sequenced as described elsewhere[16].

2.3 Gene cloning

Wild-type norM was cloned from CEP040 by PCR. Forward and reverse primers 5′-TAGAATTCATGTCGCATTCCGGTCTCA-3′ (PnorM-5′) and 5′-TCGAGATCCGGAAATTCTGCAGCGA-3′ (PnorM-3′), respectively, were designed based on the genomic DNA sequences flanking the transposon insertion site in mutant 5136. A 1666-bp product was amplified from CEP040 chromosomal DNA using the GC-Rich PCR System (Roche Diagnostics, Dorval, QC, Canada). The amplicon was treated with T4 DNA polymerase and then digested with Eco RI. This product was ligated into the vector pEX1 that had been digested with Eco RI and Sma I. The resulting construct was transformed into E. coli DH5α cells and verified by DNA sequencing. The B. vietnamiensis norM nucleotide sequence has been submitted to the GenBank nucleotide sequence database under accession No. AF312031.

2.4 Construction of plasmids for expression of norM in B. vietnamiensis

Plasmid pEX1, used to initially clone norM, contains a ColE1 origin of replication and a Ptac promoter, both of which are not functional in B. vietnamiensis. pCFG1 was constructed by replacing the tac promoter from the expression vector pKK223-3[17] with the dhfr (dihydrofolate reductase) promoter from pTnMod-OTp′[16]. This modification was necessary since we have observed that the tac promoter does not work well to drive gene expression in B. cepacia complex strains. A 308-bp product containing the dhfr promoter (Pdhfr) was amplified from pTnMod-OTp′ using forward primer 5′-CGCAACGGATCCAGAACCTTGACC-3′, reverse primer 5′-CGGAATTCGTCGAATCCTTCTTGTGAATC-3′, and Pwo DNA polymerase. This product was digested with Eco RI and ligated to plasmid pKK223–3 that had been digested with Nru I and Eco RI to remove the Ptac promoter. Plasmid pCFG1 was then digested with Eco RI and Sma I and ligated to the full-length norM as described above for pEX1. For functional complementation studies in B. vietnamiensis CEP040, norM preceded by Pdhfr was removed from pCFG1 by restriction digestion with Bam HI and subcloned into the Bam HI site of pMETpΔTet. pMETpΔTet was constructed from the Pseudomonas expression vector pME6000 as follows. pME6000 was digested with Eco RI and Pst I and ligated to the 921-bp Eco RI/Pst I fragment of pScosBC1, containing the Tp resistance gene dhfr. The resulting construct, pME6000Tp, was then digested with Eco RV and re-ligated to remove most of the tetA gene, generating pMETpΔTet.

2.5 Southern blot

Chromosomal DNA from B. cepacia complex strains was digested with Not I and the resulting fragments were separated by agarose gel electrophoresis. The DNA fragments were transferred to a positively charged nylon membrane according to the manufacturer's instructions (Roche Diagnostics). The full-length norM gene was removed from pEX1 by restriction digest with Eco RI and Hin dIII, gel purified, and labeled with digoxigenin (DIG)-UTP (Roche Diagnostics). Hybridization of the DIG-labeled norM probe was carried out at 42°C in 50% formamide for 18 h. Hybridized probe was detected using anti-DIG Ig Fab conjugated to alkaline phosphatase followed by enhanced chemiluminescence (Roche Diagnostics).

2.6 Antibiotic sensitivity assays

Mueller–Hinton broth (Difco Laboratories, Detroit, MI, USA) supplemented with varying concentrations of norfloxacin or PMB was inoculated with 10 μl of stationary phase bacterial culture. For PMB sensitivity assays on CEP040 mutants, 100 μg ml−1 Tc and/or Tp were also added as required. For norfloxacin sensitivity assays on E. coli KAM3 cells, the cultures were supplemented with 1 mM isopropyl β-d-thiogalactoside (IPTG) and 100 μg ml−1 Ap. After incubation for 18 h at 37°C the optical density of the cultures was measured at 540 nm. Survival was expressed as the ratio of the OD540 in the presence of antibiotic to that in the absence of antibiotic. All assays were carried out in triplicate and repeated at least three times.

3 Results

3.1 B. vietnamiensis CEP040 contains a norM homologue

To identify genes involved in PMB resistance, the B. vietnamiensis strain CEP040 was mutagenized with the plasposon pTnMod-OTc[16]. Mutants were screened by replica plating on LB agar containing Tc in the presence or absence of 60 μg ml−1 PMB. Thirteen PMB-sensitive mutants were obtained after screening over 50 000 Tc-resistant mutants. The sites of transposon insertion in these mutants were determined by direct cloning and sequencing of flanking host DNA sequences. By comparison of these sequences to GenBank submissions none of the insertions were located in genes previously known to be involved with PMB resistance in other Burkholderia species[18]. The characterization of the other mutants will be reported elsewhere.

The predicted amino acid sequence of the genomic DNA downstream from the TnMod-OTc insertion in the PMB-sensitive mutant 5136 exhibited significant homology to the product of the norM gene of Vibrio parahaemolyticus (GenBank accession No. O82855)[19]. NorM is a multi-drug efflux protein belonging to the MATE family of transporters[13]. Following nucleotide sequencing of genomic DNA flanking either side of the transposon, PCR primers were designed to clone the wild-type gene from B. vietnamiensis CEP040 into plasmid pEX1. The sequence of the putative CEP040 norM gene predicted a 462-amino acid protein with an estimated molecular mass of 47.9 kDa. A BLASTP search revealed strong homologies (expected values equal to or less than 7e-34) to NorM homologues from many different bacteria including Brucella melitensis, Mesorhizobium loti, Sinorhizobium meliloti, Synechocystis sp., Pseudomonas aeruginosa, Agrobacterium tumefaciens, Vibrio cholerae, Yersinia pestis, Haemophilus influenzae, Salmonella enterica LT2, and E. coli.

3.2 norM from B. vietnamiensis functionally complements the norfloxacin hypersensitivity of E. coli KAM3

To determine whether B. vietnamiensis CEP040 NorM could function as a drug efflux pump, E. coli strain KAM3 was transformed with pEX1-norM (Table 1). KAM3 cells contain a chromosomal deletion in the acrAB genes, which encode the major multi-drug efflux system of E. coli[19]. As a result, KAM3 cells are hypersensitive to norfloxacin, with a minimal inhibitory concentration of 0.03 μg ml−1[19]. As shown in Fig. 1, KAM3 cells harboring pEX1-norM (KAM3/norM) and induced with 1 mM IPTG were significantly more resistant to norfloxacin than the corresponding vector-control cells (KAM3/pEX1). Thus, B. vietnamiensis norM restores norfloxacin efflux in E. coli KAM3. In contrast, CEP040 transposon mutant 5136 was resistant to norfloxacin at concentrations greater than 125 μg ml−1 (data not shown), indicating that norM is not required for norfloxacin resistance in B. vietnamiensis.


Norfloxacin sensitivity of KAM3 transformants. E. coli KAM3 transformed with pEX1-norM (KAM3/norM) or pEX1 alone (KAM3/pEX1) were incubated overnight at 37°C in Mueller–Hinton broth supplemented with 100 μg ml−1 Ap, 1 mM IPTG and varying concentrations of norfloxacin. Survival is expressed as the ratio of the optical density at 540 nm in the presence of norfloxacin to that in the absence of norfloxacin.

3.3 norM homologues are present in representatives of six B. cepacia genomovars

To determine whether norM was unique to B. vietnamiensis, genomic DNA was prepared from isolates representing B. cepacia genomovars I–VI and analyzed by Southern blot hybridization using the CEP040 norM gene as a probe. The probe hybridized to genomic DNA from all six genomovars digested with Not I (data not shown). Thus, norM homologues are highly conserved among B. cepacia complex strains.

3.4 The effect of NorM on PMB resistance of B. vietnamiensis

Mutant 5136 was selected for study from a pool of randomly generated mutants on the basis of sensitivity to PMB. To determine whether the observed PMB sensitivity was due to disruption of norM, mutant 5136 was transformed with pMETpΔTet-norM carrying the norM gene under the control of the dhfr promoter. The Ptac promoter from pEX1 was not used in these experiments since it was inactive in B. vietnamiensis CEP040 (data not shown). Fig. 2 shows that mutant 5136(pMETpΔTet-norM) was significantly less sensitive to PMB than the vector-alone control strain over the range of 5–50 μg ml−1 PMB when the cells were grown in the presence of Tc. However, pMETpΔTet-norM did not fully complement 5136 PMB sensitivity to wild-type levels. These data could not be interpreted as a result of another mutation in mutant 5136 that contributed to PMB sensitivity, since Southern blot analysis of genomic DNA, using the tetA gene of the transposon as a probe, indicated that there was a single transposon insertion in this isolate (data not shown). Also, the product of the gene located downstream from norM is homologous to a fumarase, which is a metabolic enzyme involved in the citric acid cycle. Therefore, a polar effect on the fumarase by the transposon insertion in norM would not be expected to cause PMB sensitivity.


PMB sensitivity of CEP040 and norM mutants. Mueller–Hinton broth supplemented with varying concentrations of PMB was inoculated with B. vietnamiensis strains and incubated at 37°C. After 18 h, growth was assessed by measuring the optical density of the cultures at 540 nm. Survival is expressed as the ratio of OD540 in the presence of PMB to that in the absence of PMB. The norM transposon mutant 5136 transformed with pMETpΔTet-norM or pMETpΔTet alone was grown in the presence of Tc or Tp.

As disruption of norM was not the sole reason for increased PMB sensitivity of mutant 5136, we investigated the possibility that Tc may be involved in altering the PMB sensitivity of the mutant. Thus, we assayed PMB sensitivity of mutant 5136 transformed with pMETpΔTet-norM or vector alone in the presence and absence of Tc, and compared it to that of CEP040 (Fig. 2). In the absence of Tc there was virtually no difference in percent survival among 5136/pME, 5136/norM or CEP040. In contrast, growth of 5136/pME in the presence of Tc resulted in significant sensitivity to PMB that could be partially complemented by norM expression. These data indicate that the presence of tetA and/or Tc can alter the response of CEP040 cells to PMB, and under these conditions the NorM efflux protein can contribute to PMB resistance.

4 Discussion

In this study, we have described the cloning and characterization of the B. vietnamiensis CEP040 norM gene, and we also show that this gene is conserved in other B. cepacia complex genomovars. NorM is a member of the MATE family of transporter proteins that are widely found in a variety of species including those belonging to the Bacteria, Archaea and Eukarya kingdoms[13].

B. vietnamiensis norM conferred norfloxacin resistance to an E. coli acrAB mutant, supporting a role for NorM in drug efflux. However, the sole disruption of norM in B. vietnamiensis is not required for high levels of resistance to norfloxacin since mutant 5136 was as resistant to this drug as the parent isolate CEP040. Similarly, a norM-defective mutant in V. parahaemolyticus was resistant to fluoroquinolones, although the cloned gene were shown to be functional in E. coli for fluoroquinolone efflux[19]. Our data suggest the presence of other efflux systems for fluoroquinolones in B. vietnamiensis that may compensate for the absence of NorM. An RND-type multi-drug efflux system encoded by ceoABopcM has been described in a B. cepacia CF isolate and has been shown to confer resistance of these cells to the fluoroquinolone ciprofloxacin, as well as to Tp and chloramphenicol[14]. Also, the recently sequenced B. cepacia J2315 (genomovar III) genome (http://www.sanger.ac.uk/Projects/B_cepacia) contains at least two norM homologues (data not shown). Therefore, it is conceivable that B. vietnamiensis CEP040 may also encode additional MATE-type or RND-type efflux systems, accounting for the inherent norfloxacin resistance of this organism.

Mutant 5136 was originally selected for study based on resistance to PMB. Recent studies in other bacteria indicate that cationic peptide antibiotics can also be substrates for efflux pumps [2022]. Our results indicate that there may be a role for active efflux, in addition to low outer membrane permeability, in the resistance of B. vietnamiensis to PMB. However, the role of NorM in PMB resistance may be important only under certain growth conditions, particularly in the presence of Tc, as demonstrated by the marked PMB sensitivity to this peptide in the norM mutant that could be partially complemented by norM expression from a plasmid. At present, it is not clear how Tc can alter the PMB sensitivity of B. vietnamiensis but studies to identify the mechanism(s) involved in Tc enhancement of PMB sensitivity are currently underway in our laboratory.

Although the physiological role of NorM in B. vietnamiensis is as yet unidentified, we suggest that NorM may have some function in protection against cationic peptides, particularly when the cell is stressed in some fashion. We can speculate that this may have particular relevance for bacterial survival in the CF lung, where cells would be expected to endure many assaults including the presence of antimicrobial peptides, antibodies, and osmotic stress, as well as exogenously administered antibiotics. Alternatively, NorM may play a role in the survival of the microorganism in the soil where it may encounter various antimicrobial agents and environmental stresses.


We thank the colleagues referenced in Table 1 for the gift of strains and plasmids used in this study. This work was supported by grants from the Canadian Institutes of Health Research and the Canadian Cystic Fibrosis Foundation (to M.A.V.). C.C.F.-G. was supported by a Postdoctoral Fellowship from the Canadian Cystic Fibrosis Foundation.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
View Abstract